Methods And Systems for Detecting Variations in Minor Total-Impedance Contributors in Electrochemical Cells

This application is a continuation of U.S. patent application Ser. No. 18/066,404, filed on 2022 Dec. 15, which claims the benefit under 35 U.S.C. § 119(e) of U.S. Provisional Patent Application 63/265,480, filed on 2021 Dec. 15, which is incorporated herein by reference in its entirety for all purposes.

Determining different failure modes in electrochemical cells can be challenging. An electrochemical cell is a complex system of mechanical, electrical, and chemical components with various overlapping functions and characteristics. Once electrochemical cells are integrated into battery modules and packs, the complexity only increases. Furthermore, additional problems can be added during this integration, such as connections to the cells. At the same, the importance of early and accurate detection of various failure modes only increases.

What is needed are new methods and systems for determining failure modes attributable to variations in minor total-impedance contributors in various sets of electrochemical cells.

Described herein are methods and systems for detecting variation in minor total-impedance contributors in sets of electrochemical cells. For example, a method comprises maintaining a substantially constant current through the set of electrochemical cells and obtaining multiple voltage readings from this while the substantially constant current is maintained. The method then proceeds with determining multiple differential capacity values from the multiple voltage readings, characterizing one or more peaks in the multiple differential capacity values, and determining the variation in the minor total-impedance contributor based on one or more peaks. More specifically, partial capacitance values can be assigned to different impedance channels based on these peaks or, more specifically, based on the separation of adjacent peaks. The variation in the minor total-impedance contributor can be attributed to one or more of a tab-weld quality, electrolyte wetting, tape damage, active material activation energy variations, and diffusion variation of the ion-conducting material.

For example, non-uniform welding of one or more connections in battery cells (e.g., welding of internal tabs to electrode current collectors, tab welds to cell terminals) can contribute to different types of impedance, reflected in the total cell capacity. These different impedance types can be detected with methods disclosed herein. Specifically, different impedance types correspond to different cell capacity signatures, which can be obtained during cell charge/discharge. In other words, these different capacity signatures will reflect different impedance channels. In another example, described methods can be used to detect the activation of electrode active materials or, more specifically, when this activation is not uniform. For example, nonuniform activation will cause an impedance change of the different capacity parts and, therefore, spread out over the impedance channels. In yet another example, described methods can be used to detect nonuniform electrolyte filling, e.g., which cases different electrolyte impedances in the cell. When solid electrolytes are used, the method can detect non-uniformity in the electrolyte thicknesses and/or conductivity.

In some examples, a method of detecting a variation in a minor total-impedance contributor of a total impedance in a set of electrochemical cells is provided. The method comprises maintaining a substantially constant current through the set of electrochemical cells, obtaining multiple voltage readings from the set of electrochemical cells while the substantially constant current is maintained through the set of electrochemical cells, determining multiple differential capacity values from the multiple voltage readings, and determining the variation in the minor total-impedance contributor based on the multiple differential capacity values.

In some examples, the method further comprises characterizing one or more peaks in the multiple differential capacity values. Specifically, determining the variation in the minor total-impedance contributor is performed based on these peaks, which are associated with different impedance channels. For example, characterizing these peaks in the multiple differential capacity values can comprise one of (a) determining second-order differential capacity values from the multiple differential capacity values, or (b) comparing a plot of the multiple differential capacity values to a reference plot.

In some examples, the total impedance is determined by the minor total-impedance contributor and a major total-impedance contributor. The minor total-impedance contributor is attributed to a first portion of the set of electrochemical cells. The major total-impedance contributor is attributed to a second portion of the set of electrochemical cells, connected in series with the first portion.

In some examples, the set of electrochemical cells comprises multiple electrochemical cells connected in parallel.

In some examples, the variation in the minor total-impedance contributor is attributed to one or more characteristics selected from the group consisting of tab-weld quality, electrolyte wetting, tape damage, active-material activation energy variations material, and diffusion variations of an electrolyte of the electrochemical cells. For example, the variation in the minor total-impedance contributor is used to differentiate one of these characteristics. In more specific examples, the method further comprises associating the variation in the minor total-impedance contributor with one or more battery defects in the set of electrochemical cells.

In some examples, the multiple voltage readings are obtained from the set of electrochemical cells when the electrochemical cells are at a state of charge (SOC) selected based on and away from phase transition peaks of active materials of the electrochemical cells.

In some examples, the method further comprises (a) maintaining an additional substantially constant current through the set of electrochemical cells, wherein the additional constant current is different from the constant current, (b) obtaining additional multiple voltage readings from the set of electrochemical cells while the additional substantially constant current is maintained through the set of electrochemical cells, (c) determining additional multiple differential capacity values from the additional multiple voltage readings, (d) characterizing one or more additional peaks in the multiple additional differential capacity values, and (e) determining the variation in the minor total-impedance contributor based on theses additional peaks. For example, the additional substantially constant current is selected such that the one or more additional peaks, in the additional multiple differential capacity values, are more detectable than one or more peaks in the multiple differential capacity values determined while the set of electrochemical cells is subjected to the substantially constant current.

In some examples, maintaining the substantially constant current through the set of electrochemical cells is performed at a first temperature. The method further comprises (a) heating or cooling the set of electrochemical cells to a second temperate, different from the first temperature, (b) maintaining the substantially constant current through the set of electrochemical cells while the set of electrochemical cells is at the second temperate; (c) obtaining an additional set of multiple voltage readings from the set of electrochemical cells while the substantially constant current is maintained through the set of electrochemical cells; (d) determining additional multiple differential capacity values from the additional multiple voltage readings; and (e) determining the variation in the minor total-impedance contributor based on the additional multiple differential capacity values. In some examples, the difference between the first temperature and the second temperature is at least about 10° C. In the same or other examples, the second temperature is selected such that one or more additional peaks, in the additional multiple differential capacity values, are more detectable than one or more peaks in the multiple differential capacity values determined while the set of electrochemical cells is at the first temperature.

Also provided is an apparatus for detecting a variation in a minor total-impedance contributor of a total impedance in a set of electrochemical cells. The apparatus comprises a current source configured to flow a substantially constant current through the set of electrochemical cells, a voltmeter configured to obtain multiple voltage readings from each electrochemical cell in the set of electrochemical cells while the substantially constant current is applied to the set of electrochemical cells, and a processing element configured to (a) determine multiple differential capacity values from the multiple voltage readings, and (b) determine the variation in the minor total-impedance contributor based on the multiple differential capacity values.

In some examples, the processing element is further configured to screen the set of electrochemical cells and associate the variation in the minor total-impedance contributor with one or more battery defects in the set of electrochemical cells. In the same or other examples, the processing element is further configured to identify one or more electrochemical cells in the set of electrochemical cells when the variation in the minor total-impedance contributor associated with each of the one or more electrochemical cells is above a threshold. For example, the threshold is one of an expected-value threshold and a mean-of-population threshold. Alternatively, the threshold is the mean-of-population threshold. The electrochemical cells are identified in the set of electrochemical cells when the variation in the minor total-impedance contributor associated with each of the one or more electrochemical cells is away from the mean-of-population threshold by at least a set Z-score.

Also provided is an apparatus for in-situ diagnostics of a set of electrochemical cells based on variation in a minor total-impedance contributor of a total impedance in the set of electrochemical cells. The apparatus comprises a battery charger configured to flow a substantially constant current through the set of electrochemical cells, a battery management system configured to obtain multiple voltage readings from each electrochemical cell in the set of electrochemical cells while the substantially constant current is applied to the set of electrochemical cells, and a processing element configured to (a) determine multiple differential capacity values from the multiple voltage readings, and (b) determine the variation in the minor total-impedance contributor based on the multiple differential capacity values.

These and other embodiments are described further below with reference to the figures.

In the following description, numerous specific details are outlined to provide a thorough understanding of the presented concepts. The presented concepts may be practiced without some or all these specific details. In other instances, well-known process operations have not been described in detail to not unnecessarily obscure the described concepts. While some concepts will be described in conjunction with the specific embodiments, it will be understood that these embodiments are not intended to be limiting.

Failure modes of individual electrochemical cells, battery modules comprising multiple cells, and/or battery packs comprising multiple battery modules can take different forms.illustrates some examples of these failure modes. Some failure modes are associated with very distinctive characteristics, while others are more subtle. For example, an imperfect tab connection (e.g., insufficient weld) can cause a measurable increase in the impedance of that particular tab, but the increase in the overall cell impedance can be hard to detect using conventional techniques (e.g., impedance spectroscopy, overvoltage assessment, and the like). Specifically, with multiple tabs connected in parallel, the low impedance of the many well-formed tab connections dominates the impedance and overvoltage behavior of a single imperfect tab connection as further described below making this imperfect tab connection particularly hard to detect.

Difficulties with determining poor tab connections are associated with finding the increased impedance of the defective tab connection in parallel with the many small impedances of each well-formed tab connection. For example, a typical impedance of the welded-tab connection, which is configured to support a 60 Amp current, is about 2 mOhm. When multiple tabs are connected in parallel and one of these tabs has an increased impedance (e.g., due to an imperfect weld), the tabs with a smaller impedance will dominate the total impedance. Table 1 table illustrates a representative model in which ten resistors are connected in parallel. Specifically, in Example 1, each of these ten resistors (R-R) has an impedance of 2 mOhm, yielding a total impedance of 0.2 mOhm. In Example 2, one of these ten resistors (R) now has an impedance of 10 mOhm, while the remaining ones (R-R) are still at 0.2 mOhm. The total impedance is now 0.217 mOhm or 8.7% higher than that in Example 1. It should be noted that this change in the total impedance is minor although the impedance of Rhas increased by 400%.

Table 2 table illustrates another representative model in which ten resistors are also connected in parallel but have different impedances. Referring to Example 1, each of the nine resistors (R-R) has an impedance of 2 mOhm, while one resistor (R) has an impedance of 200 mOhm, yielding a total impedance of 0.221976 mOhm. Now, referring to Example 2, if Ris increased by 400% to 1000 mOhm, the total impedance goes up to 0.222173 mOhm, a rise of only about 0.9%. In fact, if Ris completely disconnected, which corresponds to an infinite impedance as shown in Example 3, the total impedance goes up to 0.222222 mOhm, a rise of only 0.11%. It should be noted that these impedance differences are in the order of tenths of micro-Ohms, which can be very difficult to detect and may require specialized equipment, which may not be always available. Furthermore, the impedance changes of one component can be easily masked by the impedance changes of another component, especially when the “masking” component has a significantly smaller starting resistance.

In other words, it is difficult, if possible, at all to detect changes in the overall circuit impedance when these changes are caused by changes in the impedance of one (or very few) of many components that are connected in parallel, and/or when these one or more components (with the changed impedance) are higher impedance components than one or more other components connected in parallel. It should be noted that in parallel circuits, a higher number of interconnected components reduces the contribution of each component to the total impedance. Furthermore, the contribution to the total impedance is inversely proportional to the impedance of each component,

As such, the contribution (to the total impedance) of the most resistive component can be significantly less than that of the least resistive component. More specifically, determining a resistance change of the most resistive component in a parallel connection is very difficult when using a measurement of the total resistance. For purposes of this disclosure, a circuit component that has a contribution to the total impedance of less than 50%, is referred to as a minor total-impedance contributor. In other words, if a minor total-impedance contributor is completely disconnected from the circuit, the total impedance of the circuit will change by less than 50%. This minor total-impedance contribution can result from the component being one of many resistors connected in parallel (e.g., as illustrated in Table 3) and/or from having a larger impedance than other components, to which this component is connected in parallel (e.g., as illustrated in Table 4).

The circuit models described above with reference to Tables 1-4 are representative of electrochemical cells and various battery assemblies, which are formed using these cells, as will now be described with reference to. Specifically,illustrates battery cell, which may also be referred to as a stacked cell. In this example, battery cellcomprises 3 positive electrodes and 3 negative electrodes. However, one having ordinary skill in the art would understand that any number of electrodes can be stacked together to form battery cell. Each electrode is connected to a separate tab, which is also connected to a corresponding one of the two cell terminals. In other words, this 3-electrode-pairs cell has 12 internal connections. Many modern stacked cells have up to 25 electrode pairs or even up to 50 electrode pairs, with corresponding numbers of various connections in the cells. Referring to, it should be noted that all positive-electrode tabs are connected in parallel with each other. Similarly, all negative-electrode tabs are connected in parallel with each other. The connection between the positive-electrode tabs and the negative-electrode tabs is an in-series connection provided by the electrode stack. Furthermore, the two connections provided by each tab are in series connections. In other words, the overall connection scheme is a complex combination of mechanical, electrical, and electromechanical components. This complexity makes it very difficult to determine various failure modes using externally-obtained characteristics of battery cellsuch as voltage (e.g., open-circuit voltage (OCV), constant-current voltage (CCV)), current, and the like.

Integrating electrochemical cells into a battery module only increases the complexity as will now be described with reference to. Specifically,illustrates battery assemblycomprising eighteen electrochemical cells. In addition to the internal connections formed within each cell, the cells are connected to the negative and positive terminals of the battery module.

A testable unit within a battery module depends on the ability to control the operation of this unit (e.g., to maintain a substantially constant current) and on the ability to obtain various operating characteristics specific to this unit (e.g., voltage readings). In this disclosure, a testable unit may be referred to as a set of electrochemical cells. In some examples, this set includes only one cell, e.g., as schematically shown in. In other examples, the set includes multiple cells, which can be interconnected in parallel (as, e.g., shown in), in series (as, e.g., shown in), or a combination of both interconnections.

As described above with reference to, a set of electrochemical cells can be a complex combination of mechanical, electrical, and electrochemical features, each contributing to the total impedance of the set. Furthermore, as noted above with reference to Tables 1-4, some of these features can be modeled as individual resistors. Another model will now be described with reference to. Specifically,is a model representing a testable unit comprising one or more electrochemical cells. This model is presented as a combination of partial capacities (C, C, C, and C) and partial impedances (R, R, R, and R). In this example, Ris the lowest impedance, while Ris the highest, i.e., R<R<R<R. The total capacity of this testable unit is the sum of the partial capacities (i.e., C-total=C+C+C+C). The partial capacities can have the following order C>C>C, C. One having ordinary skill in the art would recognize that various other examples are within the scope.

The current applied to the testable unit (modeled in) will cause different over-voltages at different resistors, which will cause the separation of the capacity peaks corresponding to different partial capacities as, e.g., is schematically shown in. Specifically, higher impedances will move corresponding partial capacities to the right as, e.g., is schematically shown in. Specifically, the X-axis identifies different impedance channels, while the Y-axis identifies differential capacities (dQ/dV) values.illustrates how differential capacities can be separated into distinctive impedance channels.

The integral of the Ohmic-specific capacities of the different impedance channels results in the partial capacities C, C, C, and C. It is important to note that the resistivity of the tabs is just one example of different characteristics that the described methods are capable of detecting. The methods can also be used to separate capacities accordingly to the electrolyte impedances, different activation impedances, or different diffusion impedances. In general, any impedance component of an electrochemical cell (e.g., a fuel cell, a battery cell, an analytical cell) or a capacity cell (a super-capacitor, a liquid capacitor, a solid-state capacitor) that is associated with a specific partial capacity component can be assessed by the methods described herein.

is a process flowchart corresponding to methodof detecting a variation in the minor total-impedance contributor of the total impedance in the set of electrochemical cells, in accordance with some examples. For purposes of this disclosure, a minor total-impedance contributor is defined as a resistance whose changes (e.g., an increase) have a minor effect on the total resistance. The minor effect can be defined as a change to the total resistance of less than 10% when the changes in the minor total-impedance contributor are at least 100%. The minor total-impedance contributor can correspond to the resistance of a circuit element (e.g., a battery cell (or a component of a battery cell), a tab, and a weld, active material, or electrolyte) that is connected in parallel with other components as described above. Various examples and features of minor total-impedance contributors are described above with reference to. Furthermore, various causes for this variation are shown in. Some notable examples include manufacturing errors, physical stresses applied to the cell and/or pack, and coefficient of thermal coefficient (CTE) mismatch with thermal cycling, non-uniform active material, non-uniform coating, non-uniform electrolyte thickness, and non-uniform electrolyte conductivity, e.g., any non-uniformity in electrochemical cells.

In some examples, methodcomprises (block) selecting a set of electrochemical cellsfor testing. This is an optional operation and, in some examples, all electrochemical cellscan be used in testing. For example, this test can be used during the fabrication of electrochemical cellsand/or battery packs assembled using electrochemical cells. In some examples, multiple sets of electrochemical cellscan be identified and tested in sequence. Furthermore, various other defects detection algorithms (e.g., based on current, voltage, and temperature data) of different cells in a pack in can be used to identify the cells for further testing in accordance with this method. For example, these other defect detection algorithms can identify a general problem with one or more cells, while the methods described herein can pinpoint the specific issue. For example, the set of electrochemical cellscan include one or more cells selected from a larger block of cells (e.g., a battery module or a battery pack). The remaining cells may not be tested and can be used for normal operations. For example, methodcan be performed in situ (on the set of electrochemical cells) while a battery pack remains operational and supports the external charging/discharging demands. In some examples, further described below, this in situ testing can be performed on the individual cell level. Alternatively, testing is performed on a group of multiple cells, e.g., an entire battery module. Individual cells in the tested battery module can be interconnected in accordance with the various connection schemes described above. In general, testing is performed on a testable unit, which depends on the ability to control the operation of this unit (e.g., to maintain a substantially constant current) and also on the ability to obtain various operating characteristics specific to this unit (e.g., voltage readings).

Methodproceeds with (block) maintaining a substantially constant current (or a substantially constant power) through the set of electrochemical cells. For purposes of this disclosure, the term “substantially” can be defined as no more than 25% of the peak-to-peak current variation or, more specifically, no more than 20%, no more than 15%, or even no more than 10%.

In some examples, the substantially constant current is selected such that a product of the substantially constant current and the (expected) minor total-impedance contributor is greater than a threshold set by a voltmeter (used for obtaining the multiple voltage readings in a later operation). For example, the threshold can be set by the desired width or distance of an impedance channel. For example, if a higher resolution is needed, the current can be increased to create an increased voltage differentiation from impedance to the other. Another consideration is the ability to differentiate from the main peak, provided by the major total-impedance contributor. For example, if the minor total-impedance contributor is expected to be 1 mOhm and the constant current is set to 50 A, the impedance peak is 50 mV. Of course, the current can be limited by the design of the tested unit (e.g., the size/current rating of cells, the connection among multiple cells, and the like).

For example, the total impedance is determined by a minor total-impedance contributor and a major total-impedance contributor, e.g., as a combination of these contributors corresponding to the circuit design (e.g., in-series connections, parallel connections). In more specific examples, the minor total-impedance contributor is attributed to a first portion of the set of electrochemical cells, while the major total-impedance contributor is attributed to a second portion of the set of electrochemical cells, connected in series with the first portion. IN this example, the total impedance is at least in part a sum of the minor total-impedance contributor and the major total-impedance contributor. It should be noted that differentiating of minor total-impedance contributors can be difficult because of their minor impact to the total impedance. Various methods and systems described herein are specifically configured to enable this differentiation.

Different constant current rates produce different differential capacity profiles as will now be described with reference to. Specifically,illustrates four differential capacity profiles obtained from the same cell at different constant-current rates. Linecorresponds to the smallest current. Linecorresponds to a higher current, while linecorresponds to an even higher current. Finally, linecorresponds to the highest current among these four. In this example, the current ranged from C/6 up to C/2. A higher current (line) causes the largest voltage to spread (which is evidenced by the widest and shortest peak) due to the higher over-voltages of one or more impedance components (e.g., Ohmic, activation, and/or diffusion components). As such, the capacity peaks of each electrochemical reaction are also broadened.

illustrates another example of three profiles differential capacity profiles (line, line, and line). In this example, each of the two profiles (lineand line) has a single peak (pointand point), while the third profile (line) has a dual-peak profile (pointsand). Considering the difference between these peaks (pointsand) of 100 mV and the test current of 50 A, the impedance was estimated at 2 mOhm. It should be noted that the single-peak profiles (lineand line) do not allow perform a similar impedance analysis.

are graphical representations of a testable unit model and a corresponding characterization of this model using an Ohmic-specific differential-capacity algorithm. In this characterization, the Ohmic losses (Vpp) are proportional to the cell current (i) and the impedance channels can be differentiated using the following formula:

are graphical representations of a testable unit model and a corresponding characterization of this model using an activation current-based differential-capacity algorithm. Exchange currents passing through the corresponding modeled “resistors” can be estimated by the following formulas:

In this example, the activation loss (Vpp) is dependent on log (i/i) and can be expressed as follows:

are graphical representations of a testable unit model and a corresponding characterization of this model using a current-density current-based differential-capacity algorithm, which can be represented by:

Referring to, the current densities are a*ic through Cand (1−a)*ic—through C. As such, Vpp is dependent on the nonuniform current densities log (a/(1−a)) and can be expressed by the formulas: